pith. machine review for the scientific record. sign in

arxiv: 2604.14652 · v1 · submitted 2026-04-16 · 💻 cs.RO

Recognition: unknown

DigiForest: Digital Analytics and Robotics for Sustainable Forestry

Marco Camurri , Enrico Tomelleri , Mat\'ias Mattamala , Sebasti\'an Barbas Laina , Martin Jacquet , Jens Behley , Sunni Kanta Prasad Kushwaha , Fang Nan
show 13 more authors
Nived Chebrolu Leonard Frei{\ss}muth Marvin Chayton Harms Meher V.R. Malladi Fan Yang Jonas Frey Cesar Cadena Marco Hutter Janine Schweier Kostas Alexis Cyrill Stachniss Maurice Fallon Stefan Leutenegger
Authors on Pith no claims yet

Pith reviewed 2026-05-10 11:26 UTC · model grok-4.3

classification 💻 cs.RO
keywords precision forestryautonomous roboticsforest inventoriesdecision support systemselective loggingsustainable forest managementEuropean forests
0
0 comments X

The pith

DigiForest combines autonomous robots, automated inventories, growth forecasting, and selective harvesters for precision forestry.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Forests cover roughly 40 percent of Europe's land and are central to meeting the EU's climate neutrality and biodiversity targets. The paper presents DigiForest as a new large-scale approach that applies digital technologies and autonomous robotics to improve forest management. The system rests on four components that work together: heterogeneous mobile robots collect detailed tree-level data, automated processes extract traits to create inventories, a decision support system forecasts growth to guide choices, and purpose-built autonomous harvesters perform low-impact selective logging. These elements have been tested in real forests in Finland, the UK, and Switzerland. A reader would care because the approach offers a concrete way to update current practices and address the challenges of sustainable forestry.

Core claim

DigiForest is a novel, large-scale precision forestry approach leveraging digital technologies and autonomous robotics. It is structured around four main components: autonomous, heterogeneous mobile robots (aerial, legged, and marsupial) for tree-level data collection; automated extraction of tree traits to build forest inventories; a Decision Support System for forecasting forest growth and supporting decision-making; and low-impact selective logging using purpose-built autonomous harvesters. These technologies have been extensively validated in real-world conditions in several locations, including forests in Finland, the UK, and Switzerland.

What carries the argument

The four-component integrated system of DigiForest that links robot-based data collection, automated trait extraction for inventories, a decision support system for growth forecasting, and autonomous selective harvesters.

If this is right

  • Heterogeneous robots can gather tree-level data at large scale without heavy human presence on the ground.
  • Automated trait extraction produces consistent forest inventories from robot-collected measurements.
  • The decision support system supplies growth forecasts that inform management choices aligned with long-term goals.
  • Purpose-built autonomous harvesters enable selective logging that reduces ecosystem disturbance compared with conventional methods.
  • The overall approach supports EU targets for climate neutrality and biodiversity by shifting forestry toward precise, data-driven operations.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the components prove reliable at scale, similar robot-plus-analytics setups could be adapted for forestry in other continents.
  • The marsupial robot configuration used for data collection might suggest hybrid mobility designs for other terrain-challenging environments.
  • Widespread use could shift forestry labor toward system oversight and data interpretation rather than manual fieldwork.
  • Ongoing data from the decision support system could help test how different management choices affect forest resilience under changing climate conditions.

Load-bearing premise

The four components integrate into a single scalable and reliable system whose results from limited European test sites generalize to broader sustainable forestry applications.

What would settle it

A complete end-to-end trial of the integrated DigiForest system in a new forest site that checks whether the combined data accuracy, growth forecasts, and logging outcomes match the sustainability standards observed in the original validation locations.

Figures

Figures reproduced from arXiv: 2604.14652 by Cesar Cadena, Cyrill Stachniss, Enrico Tomelleri, Fang Nan, Fan Yang, Janine Schweier, Jens Behley, Jonas Frey, Kostas Alexis, Leonard Frei{\ss}muth, Marco Camurri, Marco Hutter, Martin Jacquet, Marvin Chayton Harms, Mat\'ias Mattamala, Maurice Fallon, Meher V.R. Malladi, Nived Chebrolu, Sebasti\'an Barbas Laina, Stefan Leutenegger, Sunni Kanta Prasad Kushwaha.

Figure 1
Figure 1. Figure 1: Overview of the forest management approach proposed by DigiForest and [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: System overview. The Autonomy system executes the mission specified by a survey reference provided by a human operator. State estimation provides a consistent scene representation and dense clouds used for estimating tree traits. Forest inventory segments the point cloud data and estimates tree traits from point cloud data. The main output of the system is a forest inventory database with the main attribut… view at source ↗
Figure 3
Figure 3. Figure 3: Survey Interface: GUI for the operator, implemented on RViz using interactive [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: a) a tightly coupled depth-visual-inertial SLAM, b) a large-scale volumetric [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 4
Figure 4. Figure 4: Overall system architecture for autonomous exploration and mapping. The [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Planned trajectory before loop-closure (left) and trajectory adaptation strategies [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Local-only (blue) and global (red) frontiers in a set of submaps. The local-only [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: A block-diagram of the SDF-NMPC method. The range image is processed [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Visualization of a LiDAR-based flight among trees. The lower part of the [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: An overview of the proposed composite CBF safety filter. The method [PITH_FULL_IMAGE:figures/full_fig_p014_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Deployment of a marsupial mission, including (left to right) the walking [PITH_FULL_IMAGE:figures/full_fig_p015_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Integration of the spreadsheet output from the online forest inventory [16] [PITH_FULL_IMAGE:figures/full_fig_p016_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: The “DigiForests” dataset provides a longitudinal dataset of multiple forest [PITH_FULL_IMAGE:figures/full_fig_p018_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Overview of our panoptic segmentation approach. We use a LiDAR scan [PITH_FULL_IMAGE:figures/full_fig_p019_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Overview of the Decision Support System pipeline (adapted from Thripple [PITH_FULL_IMAGE:figures/full_fig_p020_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Individual tree data collected by team-WSL at Stein am Rhein, Switzerland [PITH_FULL_IMAGE:figures/full_fig_p021_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: The SAHA autonomous harvester. The SAHA robot is designed to perform cutting and forwarding operations at the same time, minimizing the number of steps required in the harvesting process. The lightweight design of the robot minimizes the impact on the forest floor and allows it to traverse and operate in confined spaces, which is crucial for selective thinning. An active suspension chassis with four indep… view at source ↗
Figure 17
Figure 17. Figure 17: SAHA traversability evaluation in parallel simulation. [PITH_FULL_IMAGE:figures/full_fig_p022_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: SAHA navigating through the forest with a prior point cloud map. (Top) [PITH_FULL_IMAGE:figures/full_fig_p024_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: The SAHA robot releases the cut tree after manual triggering. [PITH_FULL_IMAGE:figures/full_fig_p024_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: Map of the Stein am Rhein test site plots. The area is divided into labeled plots [PITH_FULL_IMAGE:figures/full_fig_p025_20.png] view at source ↗
Figure 21
Figure 21. Figure 21: The plot was chosen because it presented less undergrowth and high grass [PITH_FULL_IMAGE:figures/full_fig_p025_21.png] view at source ↗
Figure 21
Figure 21. Figure 21: Autonomous deployment of ANYmal in Oberwald: [PITH_FULL_IMAGE:figures/full_fig_p026_21.png] view at source ↗
Figure 22
Figure 22. Figure 22: 3D mesh reconstructed by BLK2FLY in Stein am Rhein. [PITH_FULL_IMAGE:figures/full_fig_p026_22.png] view at source ↗
Figure 23
Figure 23. Figure 23: 3D mesh reconstructed onboard by MRL drone in the field trip, Stein am [PITH_FULL_IMAGE:figures/full_fig_p027_23.png] view at source ↗
Figure 24
Figure 24. Figure 24: Exemplary semantic and instance predictions generated with our panoptic [PITH_FULL_IMAGE:figures/full_fig_p029_24.png] view at source ↗
read the original abstract

Covering one third of Earth's land surface, forests are vital to global biodiversity, climate regulation, and human well-being. In Europe, forests and woodlands reach approximately 40% of land area, and the forestry sector is central to achieving the EU's climate neutrality and biodiversity goals; these emphasize sustainable forest management, increased use of long-lived wood products, and resilient forest ecosystems. To meet these goals and properly address their inherent challenges, current practices require further innovation. This chapter introduces DigiForest, a novel, large-scale precision forestry approach leveraging digital technologies and autonomous robotics. DigiForest is structured around four main components: (1) autonomous, heterogeneous mobile robots (aerial, legged, and marsupial) for tree-level data collection; (2) automated extraction of tree traits to build forest inventories; (3) a Decision Support System (DSS) for forecasting forest growth and supporting decision-making; and (4) low-impact selective logging using purpose-built autonomous harvesters. These technologies have been extensively validated in real-world conditions in several locations, including forests in Finland, the UK, and Switzerland.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper introduces DigiForest as a novel large-scale precision forestry system that combines four components: (1) heterogeneous autonomous robots (aerial, legged, marsupial) for tree-level data collection, (2) automated extraction of tree traits to generate forest inventories, (3) a Decision Support System (DSS) for forecasting growth and aiding decisions, and (4) purpose-built autonomous harvesters for low-impact selective logging. It states that these technologies have been extensively validated in real-world forest conditions across sites in Finland, the UK, and Switzerland to advance EU climate neutrality and biodiversity objectives.

Significance. If the validations are demonstrated with quantitative evidence, the work could meaningfully advance sustainable forestry by showing how integrated robotics and digital analytics enable precise, low-impact management at scale, directly supporting policy goals for resilient ecosystems and wood-product use. The multi-robot and DSS integration offers a concrete architecture that other precision-agriculture efforts could adapt.

major comments (2)
  1. [Abstract] Abstract: the central claim that the four components 'have been extensively validated in real-world conditions' is unsupported by any performance metrics, success rates, error analyses, or methodological details. Without these, the assertions of reliability, scalability, and generalization to EU goals cannot be evaluated.
  2. [Main text] Main text (components 1-4 descriptions): no quantitative results are supplied for robot navigation in dense canopy, trait-extraction accuracy, DSS forecast error, or harvester impact/efficiency, even though these are required to substantiate the 'extensive validation' and integration claims.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. We agree that the current version does not provide sufficient quantitative evidence to fully substantiate the claims of extensive real-world validation, and we will revise the paper to include key performance metrics, error analyses, and references to supporting studies while preserving its role as a high-level system overview.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the four components 'have been extensively validated in real-world conditions' is unsupported by any performance metrics, success rates, error analyses, or methodological details. Without these, the assertions of reliability, scalability, and generalization to EU goals cannot be evaluated.

    Authors: We agree that the abstract's phrasing requires stronger substantiation. This manuscript is intended as an integrative overview of the DigiForest architecture rather than a detailed empirical study. The real-world validations across Finland, the UK, and Switzerland have produced quantitative results (navigation success rates, trait-extraction accuracies, DSS forecast errors, and harvester impact metrics), but these appear in companion technical papers. We will revise the abstract to state that the components 'have been validated through real-world deployments' and add a concise summary of representative metrics with citations to the detailed studies. revision: yes

  2. Referee: [Main text] Main text (components 1-4 descriptions): no quantitative results are supplied for robot navigation in dense canopy, trait-extraction accuracy, DSS forecast error, or harvester impact/efficiency, even though these are required to substantiate the 'extensive validation' and integration claims.

    Authors: We acknowledge that the main-text descriptions of the four components focus on system design and integration without embedding the supporting quantitative data. To address this, we will add a dedicated validation subsection (or expanded paragraphs within each component section) that reports key metrics from the field campaigns, including robot navigation performance in dense canopy, automated trait-extraction accuracy and error rates, DSS growth-forecast errors, and harvester efficiency/impact measures. These additions will be accompanied by references to the underlying studies and will directly support the integration and scalability claims. revision: yes

Circularity Check

0 steps flagged

No circularity: descriptive project overview with no derivations or self-referential predictions

full rationale

The manuscript is a high-level description of the DigiForest project and its four components (heterogeneous robots, trait extraction, DSS, autonomous harvesters) plus real-world validation sites. No equations, fitted parameters, predictions, or derivation chains appear in the abstract or the supplied text. Claims rest on external empirical validation rather than any reduction of outputs to the paper's own inputs by construction, self-citation load-bearing, or ansatz smuggling. This is the expected non-circular outcome for an overview paper without mathematical content.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a high-level system description paper with no mathematical derivations, fitted parameters, or postulated entities; the abstract introduces no free parameters, axioms, or invented entities.

pith-pipeline@v0.9.0 · 5588 in / 1172 out tokens · 35433 ms · 2026-05-10T11:26:56.811467+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

49 extracted references · 27 canonical work pages · 1 internal anchor

  1. [1]

    Forest Inventory with Terrestrial LiDAR: A Compar- ison of Static and Hand-Held Mobile Laser Scanning

    S´ebastien Bauwens et al. “Forest Inventory with Terrestrial LiDAR: A Compar- ison of Static and Hand-Held Mobile Laser Scanning”. In:Forests7.6 (2016). 127.issn: 1999-4907.doi:10.3390/f7060127

    work page doi:10.3390/f7060127 2016

  2. [2]

    SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences

    J. Behley et al. “SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences”. In:Intl. Conf. on Computer Vision (ICCV). 2019

    2019

  3. [3]

    Besl and Neil D

    Paul J. Besl and Neil D. McKay. “A Method for Registration of 3-D Shapes”. In:IEEE Trans. Pattern Anal. Mach. Intell.14.2 (1992), pp. 239–256.doi: 10.1109/34.121791

    work page doi:10.1109/34.121791 1992

  4. [4]

    In: IEEE International Conference on Robotics and Automation, ICRA 2024, Yokohama, Japan, May 13-17, 2024

    Simon Boche, Sebasti ´an Barbas Laina, and Stefan Leutenegger. “Tightly- Coupled LiDAR-Visual-Inertial SLAM and Large-Scale Volumetric Occupancy Mapping”. In:IEEE Int. Conf. Robot. Autom. (ICRA). 2024, pp. 18027–18033. doi:10.1109/ICRA57147.2024.10610460

    work page doi:10.1109/icra57147.2024.10610460 2024

  5. [5]

    A Simplified Forest Model to Study Species Compo- sition Along Climate Gradients

    Harald K. M. Bugmann. “A Simplified Forest Model to Study Species Compo- sition Along Climate Gradients”. In:Ecology77.7 (1996), pp. 2055–2074.doi: https://doi.org/10.2307/2265700 . eprint: https://esajournals. onlinelibrary.wiley.com/doi/pdf/10.2307/2265700 .url: https: DigiForest: Digital Analytics and Robotics for Sustainable Forestry 31 / / esajournal...

    work page doi:10.2307/2265700 1996

  6. [6]

    Extracting individual trees from lidar point clouds using treeseg

    Andrew Burt, Mathias Disney, and Kim Calders. “Extracting individual trees from lidar point clouds using treeseg”. In:Methods in Ecology and Evolution 10.3 (2019), pp. 438–445.doi: https : / / doi . org / 10 . 1111 / 2041 - 210X.13121

    2019

  7. [7]

    Zhao, Chenfeng Xu, Chen Tang, Chenran Li, Mingyu Ding, Masayoshi Tomizuka, and Wei Zhan

    Benoˆıt Casseau et al. “Markerless Aerial-Terrestrial Co-Registration of Forest Point Clouds using a Deformable Pose Graph”. In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2024, pp. 39–46.doi: 10.1109/IROS58592.2024. 10802448

    work page doi:10.1109/iros58592.2024 2024

  8. [8]

    Nived Chebrolu and Maurice Fallon.Report defining data format and software API for multi-robot map sharing. Tech. rep. University of Oxford, 2023

    2023

  9. [9]

    4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks

    C. Choy, J. Gwak, and S. Savarese. “4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks”. In:IEEE Int. Conf. Computer Vision and Pattern Recognition (CVPR). 2019

    2019

  10. [10]

    2021.url: https://eur- lex.europa.eu/legal- content/ EN/TXT/?uri=CELEX%3A52021DC0572

    European Commission.Commission communication: New EU forest strategy for 2030. 2021.url: https://eur- lex.europa.eu/legal- content/ EN/TXT/?uri=CELEX%3A52021DC0572

    2030

  11. [11]

    Steering kinematics for a center-articulated mobile robot

    P.I. Corke and P. Ridley. “Steering kinematics for a center-articulated mobile robot”. In:IEEE Transactions on Robotics and Automation17.2 (2001), pp. 215–218.doi:10.1109/70.928568

    work page doi:10.1109/70.928568 2001

  12. [12]

    Lidar Remote Sensing for Forestry

    Ralph O. Dubayah and Jason B. Drake. “Lidar Remote Sensing for Forestry”. In:Journal of Forestry98.6 (June 2000), pp. 44–46.issn: 0022-1201.doi: 10.1093/jof/98.6.44

    work page doi:10.1093/jof/98.6.44 2000

  13. [13]

    Global aboveground biomass product validation best practices protocol. Version 1.0

    L Duncanson et al. “Global aboveground biomass product validation best practices protocol. Version 1.0”. In:Best practice protocol for satellite derived land product validation. Land Product Validation Subgroup (Working Group on Calibration and Validation, Committee on Earth Observation Satellites) (2020)

    2020

  14. [14]

    A density-based algorithm for discovering clusters in large spatial databases with noise

    M. Ester et al. “A density-based algorithm for discovering clusters in large spatial databases with noise.” In:Proc. of the Conf. on Knowledge Discovery and Data Mining (KDD). 1996

    1996

  15. [15]

    On-Manifold Preintegration for Real-Time Visual– Inertial Odometry

    Christian Forster et al. “On-Manifold Preintegration for Real-Time Visual– Inertial Odometry”. In:IEEE Trans. Robot.33.1 (2017), pp. 1–21.doi: 10.1109/TRO.2016.2597321

    work page doi:10.1109/tro.2016.2597321 2017

  16. [16]

    Online Tree Reconstruction and Forest Inventory on a Mobile Robotic System

    Leonard Freißmuth et al. “Online Tree Reconstruction and Forest Inventory on a Mobile Robotic System”. In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2024

    2024

  17. [17]

    Trajectory Optimization and Following for a Three Degrees of Freedom Overactuated Floating Platform

    Jonas Frey et al. “Locomotion Policy Guided Traversability Learning using Volumetric Representations of Complex Environments”. In:2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). 2022, pp. 5722–5729.doi:10.1109/IROS47612.2022.9982190

    work page doi:10.1109/iros47612.2022.9982190 2022

  18. [18]

    In: 2023 IEEE/RSJ Interna- tionalConferenceonIntelligentRobotsandSystems(IROS).pp.7742–7749(2023)

    Lanke Frank Tarimo Fu, Nived Chebrolu, and Maurice Fallon. “Extrinsic Calibration of Camera to LIDAR Using a Differentiable Checkerboard Model”. 32 Authors Suppressed Due to Excessive Length In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2023, pp. 1825–1831.doi: 10.1109/IROS55552.2023.10341781

    work page doi:10.1109/iros55552.2023.10341781 2023

  19. [19]

    Multi-resolution 3D mapping with explicit free space representation for fast and accurate mobile robot motion planning

    Nils Funk et al. “Multi-resolution 3D mapping with explicit free space representation for fast and accurate mobile robot motion planning”. In:IEEE Robot. Autom. Lett.6.2 (2021), pp. 3553–3560

    2021

  20. [20]

    Unified temporal and spatial calibration for multi-sensor systems

    Paul Furgale, Joern Rehder, and Roland Siegwart. “Unified temporal and spatial calibration for multi-sensor systems”. In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2013, pp. 1280–1286.doi:10.1109/IROS.2013.6696514

    work page doi:10.1109/iros.2013.6696514 2013

  21. [21]

    In- formed RRT*: Optimal sampling-based path planning focused via direct sampling of an admissible ellipsoidal heuristic

    Jonathan D Gammell, Siddhartha S Srinivasa, and Timothy D Barfoot. “In- formed RRT*: Optimal sampling-based path planning focused via direct sampling of an admissible ellipsoidal heuristic”. In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2014, pp. 2997–3004

    2014

  22. [22]

    G. J. Hamilton.Forest Mensuration Handbook. 1988

    1988

  23. [23]

    Motion Primitives Planning For Center-Articulated Vehicles

    Jiangpeng Hu et al. “Motion Primitives Planning For Center-Articulated Vehicles”. In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2024, pp. 12702– 12709

    2024

  24. [24]

    Force control for active chassis balancing

    Marco Hutter et al. “Force control for active chassis balancing”. In:IEEE/ASME Transactions On Mechatronics22.2 (2016), pp. 613–622

    2016

  25. [25]

    Harveri : A Small (Semi-)Autonomous Precision Tree Harvester

    Edo Jelavic et al. “Harveri : A Small (Semi-)Autonomous Precision Tree Harvester”. In:ICRA 2022 Workshop in Innovation in Forestry Robotics: Research and Industry Adoption. 2022.url: https://openreview.net/ forum?id=BOdgBn6OQM5

    2022

  26. [26]

    iSAM2: Incremental smoothing and mapping using the Bayes tree

    Michael Kaess et al. “iSAM2: Incremental smoothing and mapping using the Bayes tree”. In:Int. J. Robot. Res.31.2 (2012), pp. 216–235.doi: 10.1177/ 0278364911430419

    2012

  27. [27]

    Panoptic Segmentation

    A. Kirillov et al. “Panoptic Segmentation”. In:IEEE Int. Conf. Computer Vision and Pattern Recognition (CVPR). 2019

    2019

  28. [28]

    OKVIS2: Realtime scalable visual-inertial SLAM with loop closure,

    Stefan Leutenegger. “Okvis2: Realtime scalable visual-inertial slam with loop closure”. In:arXiv preprint arXiv:2202.09199(2022)

    work page arXiv 2022

  29. [29]

    DigiForests: A Longitudinal LiDAR Dataset for Forestry Robotics

    Meher V.R. Malladi et al. “DigiForests: A Longitudinal LiDAR Dataset for Forestry Robotics”. In:IEEE Int. Conf. Robot. Autom. (ICRA). 2025

    2025

  30. [30]

    Mask-Based Panoptic LiDAR Segmentation for Au- tonomous Driving

    R. Marcuzzi et al. “Mask-Based Panoptic LiDAR Segmentation for Au- tonomous Driving”. In:IEEE Robot. Autom. Lett.8.2 (2023), pp. 1141–1148

    2023

  31. [31]

    An Efficient Locally Reactive Controller for Safe Navigation in Visual Teach and Repeat Missions

    Mat´ıas Mattamala, Nived Chebrolu, and Maurice F. Fallon. “An Efficient Locally Reactive Controller for Safe Navigation in Visual Teach and Repeat Missions”. In:IEEE Robot. Autom. Lett.7.2 (2022), pp. 2353–2360.doi: 10.1109/LRA.2022.3143196

    work page doi:10.1109/lra.2022.3143196 2022

  32. [32]

    Building Forest Inventories With Autonomous Legged Robots—System, Lessons, and Challenges Ahead

    Mat´ıas Mattamala et al. “Building Forest Inventories With Autonomous Legged Robots—System, Lessons, and Challenges Ahead”. In:IEEE Transactions on Field Robotics2 (2025), pp. 418–436.doi:10.1109/TFR.2025.3583972

    work page doi:10.1109/tfr.2025.3583972 2025

  33. [33]

    Simultaneous Localization and Mapping for Forest Harvesters

    Mikko Miettinen et al. “Simultaneous Localization and Mapping for Forest Harvesters”. In:Proceedings 2007 IEEE International Conference on Robotics and Automation. 2007, pp. 517–522.doi:10.1109/ROBOT.2007.363838. DigiForest: Digital Analytics and Robotics for Sustainable Forestry 33

    work page doi:10.1109/robot.2007.363838 2007

  34. [34]

    Trajectory Optimization and Following for a Three Degrees of Freedom Overactuated Floating Platform

    Takahiro Miki et al. “Elevation Mapping for Locomotion and Navigation using GPU”. In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2022, pp. 2273– 2280.doi:10.1109/IROS47612.2022.9981507

    work page doi:10.1109/iros47612.2022.9981507 2022

  35. [35]

    doi:10.1126/scirobotics

    Takahiro Miki et al. “Learning Robust Perceptive Locomotion for Quadrupedal Robots in the Wild”. In:Sci. Robot.7.62 (2022).doi: 10.1126/scirobotics. abk2822

    work page doi:10.1126/scirobotics 2022

  36. [36]

    Learning adaptive controller for hydraulic machinery automation

    Fang Nan and Marco Hutter. “Learning adaptive controller for hydraulic machinery automation”. In:IEEE Robot. Autom. Lett.(2024)

    2024

  37. [37]

    Zhao, Chenfeng Xu, Chen Tang, Chenran Li, Mingyu Ding, Masayoshi Tomizuka, and Wei Zhan

    Haedam Oh et al. “Evaluation and Deployment of LiDAR-based Place Recog- nition in Dense Forests”. In:IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS). 2024, pp. 12824–12831.doi:10.1109/IROS58592.2024.10801297

    work page doi:10.1109/iros58592.2024.10801297 2024

  38. [38]

    Efficient Submap-based Autonomous MA V Exploration using Visual-Inertial SLAM Configurable for LiDARs or Depth Cameras

    Sotiris Papatheodorou et al. “Efficient Submap-based Autonomous MA V Exploration using Visual-Inertial SLAM Configurable for LiDARs or Depth Cameras”. In:arXiv preprint arXiv:2409.16972(2024)

    work page arXiv 2024

  39. [39]

    Significant increase in natural disturbance impacts on European forests since 1950

    Marco Patacca et al. “Significant increase in natural disturbance impacts on European forests since 1950”. In:Global Change Biology29.5 (2023), pp. 1359–1376.doi:https://doi.org/10.1111/gcb.16531

    work page doi:10.1111/gcb.16531 1950

  40. [40]

    The Resilient Micro Flyer: a New Collision-tolerant Au- tonomous Robot

    Paolo De Petris. “The Resilient Micro Flyer: a New Collision-tolerant Au- tonomous Robot”. MA thesis. University of Nevada, Reno, 2020

    2020

  41. [41]

    Mapping forests using an unmanned ground vehicle with 3D LiDAR and graph-SLAM

    Marek Pierzcha la, Philippe Gigu`ere, and Rasmus Astrup. “Mapping forests using an unmanned ground vehicle with 3D LiDAR and graph-SLAM”. In: Computers and Electronics in Agriculture145 (2018), pp. 217–225.issn: 0168-1699.doi:https://doi.org/10.1016/j.compag.2017.12.034

    work page doi:10.1016/j.compag.2017.12.034 2018

  42. [42]

    Seamless Scene Segmentation

    L. Porzi et al. “Seamless Scene Segmentation”. In:IEEE Int. Conf. Computer Vision and Pattern Recognition (CVPR). 2019

    2019

  43. [43]

    Sundaresan, J

    Milad Ramezani et al. “Online LiDAR-SLAM for Legged Robots with Robust Registration and Deep-Learned Loop Closure”. In:IEEE Int. Conf. Robot. Autom. (ICRA). 2020, pp. 4158–4164.doi: 10.1109/ICRA40945.2020. 9196769

    work page doi:10.1109/icra40945.2020 2020

  44. [44]

    U-Net: Convolutional Networks for Biomedical Image Segmentation

    O. Ronneberger, P.Fischer, and T. Brox. “U-Net: Convolutional Networks for Biomedical Image Segmentation”. In:Proc. of the Medical Image Computing and Computer-Assisted Intervention (MICCAI). 2015.url: https://arxiv. org/pdf/1505.04597.pdf

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  45. [45]

    Automatic three-dimensional mapping for tree diameter measurements in inventory operations

    Jean-Franc ¸ois Tremblay et al. “Automatic three-dimensional mapping for tree diameter measurements in inventory operations”. In:Journal of Field Robotics 37.8 (2020), pp. 1328–1346.doi: https://doi.org/10.1002/rob.21980

    work page doi:10.1002/rob.21980 2020

  46. [46]

    NeRF-Supervision: Learning Dense Object Descriptors from Neural Radiance Fields , booktitle =

    Kavisha Vidanapathirana et al. “LoGG3D-Net: Locally Guided Global De- scriptor Learning for 3D Place Recognition”. In:IEEE Int. Conf. Robot. Autom. (ICRA). 2022, pp. 2215–2221.doi:10.1109/ICRA46639.2022.9811753

    work page doi:10.1109/icra46639.2022.9811753 2022

  47. [47]

    VILENS: Visual, Inertial, Lidar, and Leg Odometry for All-Terrain Legged Robots

    David Wisth, Marco Camurri, and Maurice Fallon. “VILENS: Visual, Inertial, Lidar, and Leg Odometry for All-Terrain Legged Robots”. In:IEEE Trans. Robot.39.1 (2023), pp. 309–326.doi:10.1109/TRO.2022.3193788

    work page doi:10.1109/tro.2022.3193788 2023

  48. [48]

    Unifying flow, stereo and depth estimation

    Haofei Xu et al. “Unifying flow, stereo and depth estimation”. In:IEEE Trans. Pattern Anal. Mach. Intell.45.11 (2023), pp. 13941–13958. 34 Authors Suppressed Due to Excessive Length

    2023

  49. [49]

    Angelos Zacharia, Mihir Dharmadhikari, and Kostas Alexis.Collaborative Exploration with a Marsupial Ground-Aerial Robot Team through Task- Driven Map Compression. 2025. arXiv: 2509.07655 [cs.RO].url: https: //arxiv.org/abs/2509.07655

    work page arXiv 2025